I have the KR4 in both E21A 3500k and 2200k. I love both, but at nighttime the 2200k is the one I reach for 95% of the time. The amber glow from those emitters makes the 3500k E21A look like cold white. It’s incredible. The red rendering is good enough in my opinion, because 2200k is such a warm color temperature it naturally has a lot of red in it even with only R9050, but still I really wish Hank went with the superior E21A 2000k (tested at R9080 by several people) because if R9050 2200k looks good then the 2000k must be god-like. For what it’s worth, though, the E21A 2200k at least makes reds and wood tones look slightly richer than my SST-20 2700k lights do. I think you won’t regret it if you buy it.
I have a KR4 with E21A 2200K. If you like warm tints (like I do), you will like E21A 2200K. Once my eyes adjust, the color rendering looks really good, even though R9 is very likely below 80.
Other warm emitters that you might be able to get if you ask Hank:
SST-20 2700K JA3 CRI95
XP-L HI T6 8B 2850K CRI90
Has anyone measured R9 for these yet?
Ive built a light using the 2000k. Not sure about Godlike, i dont think you can understand quite how orange it is until you see it.
Its unique and fun, but its too far, almost fluro orange. I changed the light over to a 2700k and its much better.
I would like to see the 2200k though as something inbetween would be nice. After seeing the 2000k, i really dont think i could notice the difference on the R9050, most colours have changed so much at that CCT.
Im sure ill get Hank’s 2200k at some point as if he keep releasing new lights with E21a im going to have to keep buying them and there are only so many more 3500k lights i can have!
PLEASE Hank, 3000k & 2700k next time, pretty please!
I cant imagine people in the know actually wants the 5000k unless they brought the light by mistake not knowing what an E21a is 
(Not that its “bad”, i just cant imagine tint snobs wanting it).
People love Pink 4500k, we know that, but i NEVER reach for my 219b 4500k lights, never. 2700k to 3500k R9080 FTW
Many thanks to all that answered. I ordered the KR4 Ti E21A 2200K.
If anyone has or will have CRI and R9 data on it, I am still very interested to see it.
Hank, I have been using the firmware update kit to keep the firmware of my D4V2 Ti lights up to date. I am using the normal version for my XP-L HI one and the 219 version for the ones I have modded to Nichia 219b sw45k and 219b sw30.
For the KR4 E21A, the No FET version should be used, right?
To those interested in the KR4 16x E21A mule:
- it's available now if you ask via email (thanks, Hank!)
- price is +$20 over the standard KR4 E21A
- about ~3000 Lumens max (probably not for very long :fire:)
- features a 7.5A constant current driver
I don’t think the 2000K E21A looks like any fluorescent I’ve seen in person. The closest I’ve seen would probably be this purpose built 2000K fluorescent. But that was made to specifically be 2000K, and it was built for experiments. My example of the E21A looks much closer to a candle albeit slightly lower Duv.
I think when he says “fluoro” orange he just means it is bright orange.
Yes, no FET version should be used for the E21A LED.
Thank you for verifying Hank.
Are you planning to get any other E21A tints?
Thanks!
It’s worth noting that every “9050” Nichia emitter I’ve seen a test of has been well above 50 for the R9. That’s just the minimum guaranteed rating.
Yes, I have read that, but it would be interesting to know how much more is Hank’s batch.
The E21A 2700K is actually just on the BBL so appears the same or yellower than the Oslon Square 92CRI or SST-20 2700K depending on current level. The 3000K is a bit below the BBL but nothing to write home about and I hardly even use it. The E21A I use by far the most is the E21A 2000K at night because of its almost zero blue wavelength, minimal affect on night vision, is very cozy and almost matches the ambience of a candle which is 1800K.
The first time I use the 2000K, I was underwhelmed because of how orange it is. I was like yuck. But that was because it was daytime and my eyes are not adjusted. At night time once your eyes are adapted to the 2000K temperature, it looks very nice and color rendering excellent. In fact once your eyes are adapted to the 2000K temperature and you blast on a 4000K light, it looks yucky cold blue. That’s how our eyes work. It takes time to adapt to different color temps.
That depends on whether you want it to be yellow or just warm white. Usually “white” is considered to be about 2700K to 6500K, with anything below being too yellow or red to look white, and anything above being too blue or violet to appear white.
Do you have a resource for this?
This brings up a good point. People are forever wanting their lights to “compare” to something or be more natural in some way yet the most sought after CCT on the forum seems to be around 4500K which is probably the least common CCT to find anywhere (Sun, Any Indoor Lighting of any type, flames, etc) and that is leaving DUV out of the discussion.
Here’s some food for thought. Two measurements taken in Texas, today at 2:45PM, overcast day. First is indoors with no artificial light, only ambient light from window ‘with warm, sandy gray’ wall paint - the next measurement is when I step outside.
I guess the point is there is no point. There is either too many use cases, personal taste, eye variation, ambient light to really make any judgement towards what someone should pick.
Can you believe I sit in my home office with 92 CRI and 59 R9 value? 
I’m a heathen.
It’s not a hard and fast rule, but I dont think most people would qualitatively consider candlelight temps (2000K or below I believe?) or 8000K+ as “white”. The spectral composition is very different at the extreme ends, being mostly red/orange and blue/violet respectively.
I think a big part of why 4500K is so loved is because for many it is the “perfect average”. Against daylight, sunrise/set, incandescents, florescent area lighting, etc. it never looks too out of place. 4000K or less can look too warm with any ambient sunlight, and 5000K can be harsh and sterile at night or with incandescent/sodium ambient lighting. 4000K vs 5000K seems to be biggest visual change from only 1000K difference, and it falls right in the center, seemingly having the benefits of both and less of the potential negative.
219B SW45s now legendary reputation almost certainly play a part in that too.
Oh and an anecdote. The fairly old florescent bulbs in my garage seem to be almost exactly 4500K. Only light source I’ve ever seen that seemed to be at that CCT other than flashlights.
Seemingly popular opinion does not override science. Most people have little idea what CCT is or how white light is even defined. Even within my anecdotal experience, I have a several friends that think 5000K is white light and any CCT below that is yellow.
Wow, thanks for your input, did not know the Kruithof curve, directly googled it (He was Dutch as well)
Well, for one reference, there’s ANSI standard C78.377-2008, a.k.a. “ANSI White”…
… but this stuff is really about personal preference and perception. The standard just formalizes the most common human perceptions, with 2700K at one end of the popular spectrum and 6500K at the other. Not everyone prefers lights within that range, but most of the population does.
Basically, there’s a point around roughly 4500K to 5000K which is considered a true neutral white, and the farther a light gets from that, the less white it looks. However, the color space is curved… so there is a curved line through the space which looks more white than anything above or below the line. And the question of what is “white” depends on who you ask. Each person has a different center point on the graph, and different tolerances for how far away from that point a light can get before it no longer looks white to them.
Here’s how that looks in the CIE 1960 Uniform Color Space:
